Intended for healthcare professionals

Education And Debate

Spectre of racism in health and health care: lessons from history and the United States

BMJ 1998; 316 doi: (Published 27 June 1998) Cite this as: BMJ 1998;316:1970
  1. Raj Bhopal, professor (r.s.bhopal{at}
  1. Department of Epidemiology and Public Health, Medical School, University of Newcastle, Newcastle upon Tyne NE2 4HH
  • Accepted 3 March 1998

Inequalities in health and health care in relation to race and ethnicity pose ethical problems of which racism is the most disquieting. 1 2 One controversial inequality is the poor health of African Americans—their life expectancy in 1993 was 7.1 years less than that of white Americans. La Veist et al have shown that the disparity has increased over this century.3 The deficit arises from excess mortality in relation to many causes of death, and is partly explained by differences in income.4


Race—The group a person belongs to as a result of a mix of physical features, ancestry, and geographical origins, as identified by others or, increasingly, as self identified. The importance of social factors in the creation and perpetuation of racial categories has led to a broadening of the concept to include social and political heritage, making its usage similar to ethnicity. Race and ethnicity are increasingly used synonymously

Ethnicity—The group a person belongs to as a result of a mix of cultural factors, including language, diet, religion, ancestry, and race

Racism—A belief that some races are superior to others, used to devise and justify actions that create inequality between racial groups

White—People with European ancestral origins known in the 19th century as caucasoid (in the United States, white includes people from the Middle East and north Africa) Black—People with African ancestral origins (in the United States, excluding some parts of north Africa) and who fall into the racial group known in the 19th century as negroid

African American—People who fall into the category black and live in the United States

Empirical evidence is scarce and hard to interpret, but much public opinion and some scholarly analysis in the United States and the United Kingdom place racism at the hub of ethnic and racial inequalities in health and health care. 2 3 511 Health professions, governed by ethical codes that emphasise their humanitarian duties, find the charge that health care is racist hard to bear. It is less vocal in the United Kingdom than in the United States, but the issues are similar enough for us to learn from the experience there. This essay takes a historical perspective in order to disentangle the argument and counter argument that characterise the current debate. I have used the terms found in most of the publications cited—my understanding of these terms used is given in the box.

Summary points

Inequalities in the health and health care of ethnic and racial minority groups are evident

Racism is the most disturbing of the explanations for these inequalities

Pinpointing the specific role of racism is difficult

The history of racism in science and medicine shows that people and institutions behave according to the ethos of their times and warns of dangers to avoid in the future

Inequalities result from inextricably linked, complex factors including historical and current racism

Action to reverse inequalities should not have to wait for reliable answers to questions on causes and mechanisms

History of scientific racism

Hippocrates contrasted the feebleness of the Asiatic races to the hardiness of the Europeans.12 Hippocrates' concept of race was of human groups shaped by ancestry in different geographical conditions, especially climate. In the 19th century, racial differences in anatomical, physiological, behavioural, and health status were avidly sought.1316 The idea of races as distinct species gave way to that of races as biological subspecies. This is the defining feature of the 20th century concept of race, supported by many dictionaries and encyclopaedias and permeating biomedical thinking. However, the view that race is a social and not a biological reality is emergent.

That some races are superior to others, a tenet of racism, was widely believed, especially in the 19th century,1720 but is clear in Hippocratic writings.12 In the 19th century, differences between races were usually assumed to be biological, were interpreted to show the superiority of white races, and were used to justify policies that subordinated “coloured” groups.1322 These policies included slavery,20 social inequality, 15 19 eugenics,21 immigration control, 15 21 and the unequal practice of medicine. 22 23 John Down's theory of “mongolism” (trisomy 21 or Down's syndrome) was that these infants were from an inferior, Mongoloid, race.24


In May 1997, President Clinton apologised to the survivors of the Tuskegee experiment—black men deceived and bribed into cooperating with research into the progression of syphilis

Dr J M Smith argued that the environment was responsible for health disparities, showing that poor white families had patterns of diseases similar to black ones.9 Challenges such as this are still needed. For example, the suggestion that the higher prevalence of hypertension in black people is caused by biological factors is challenged only rarely with the alternative view that it is a response to racism or other environmental factors. 58 26 27

Modern genetics undermined the biological concept of race,14 and Nazi racism undermined eugenics.28 Races are now considered to be based on a few physical features (such as colour and facial features) that are of little direct importance to health but serve important social purposes. 2 5 22 26 27 None the less, the idea of the biological basis of health differences in relation to race 26 27 and ethnicity29 remains strong.

Gamble has argued that the Tuskegee syphilis study has left a legacy of mistrust.30 This study in Alabama lasted from 1932-72, and deceived and bribed 600 black subjects into cooperating with research that examined the progression of syphilis without treatment, even once a cure was available.31 The study was conducted by the US Public Health Service with the backing of the medical and scientific establishment. In May 1997, President Clinton apologised, on behalf of America, to the survivors of this experiment. Tuskegee was not a unique racist medical experiment.30

Contemporary research in the US

Osborne concluded that much American health research on race and ethnicity is itself racist by contributing to the idea that some human groups are inferior.32 The review of racism, sexism, and class by Krieger et al concluded that racial/ethnic differences in health have not been explained.5 They criticised research on racial differences, particularly as racism—which they defined as “an oppressive system of racial relations, justified by ideology, in which one social group benefits from dominating another”—was rarely studied. Krieger et al contended that much modern research supported the assumptions needed to justify racism. They explained that racism is important because it leads to socioeconomic inequalities that underlie health inequalities, and, by implication, that racism underlies unexplained inequities in health care, including treatment for heart disease,33 renal failure,34 bladder cancer,35 and pneumonia.36 These inequalities have been documented in numerous studies.

Gornick et al, for example, showed that black people had fewer mammograms, immunisations, and ambulatory care visits than white people, but greater mortality and more admissions to hospital.37 Commenting on this, Geiger wrote that “investigators tend to invoke unspecified cultural differences, undocumented patient preferences, or a lack of information about the need for care as reasons for the differences. The alternative explanation is racism, that is, racially discriminatory rationing by physicians and health care institutions. We don't yet know enough to make that charge definitively.”38

The consistent and repeated findings that black Americans receive less health care than white Americans—particularly where this involves expensive new technology—is an indictment of American health care.39 These inequalities are not wholly a result of differences in socioeconomic circumstances. 4 5 38 39 Escarce et al explained their finding that white patients were more likely than black patients to receive services in terms of the following factors: different disease patterns; different level of contact with doctors, especially specialists; financial and organisational barriers; patients' preferences; and the fact that doctors managed their patients differently on the basis of race.40

The difficulty in interpreting these findings is considered in the context of heart disease, which has been studied in detail. Differences between black and white patients have been publicised since 1984.33 As the box shows, white patients in the United States receive more intensive medical attention in the treatment of heart disease than do black patients.

Racial inequalities in treating heart disease

  • Wenneker and Epstein showed that black patients had lower rates for coronary angiography and coronary artery bypass grafting than white patients after adjustment for confounding factors41

  • Hannan et al showed that black patients had fewer cardiac procedures than white ones after adjustment for disease severity42

  • Goldberg et al reported that coronary artery bypass rates in black men and women were a quarter and a third respectively of rates in white men and women43

  • Whittle et al showed inequalities between black and white patients in invasive cardiac procedures44

  • Ayanian et al reported that black patients had fewer coronary revascularisation procedures than white patients45

  • Carlisle et al showed that in Los Angeles invasive cardiac procedures were less common in Latin American and black patients than in white patients, but not in Asian patients compared with white patients46

  • Peterson et al showed that black patients had fewer cardiac procedures yet better short term survival and equivalent intermediate survival rates47

Thus, extensive published reports do not yield a clear conclusion on the role of racism. The studies are mainly on quantity, not quality, of care, but Ayanian et al were not clear whether white patients had too many interventions or black patients too few.45 A study examining outcomes showed no differences between black and white patients, and thus evoked a different reaction from studies examining quantity of care.47 Despite at least 15 years of attention and much research, no definitive explanation has emerged. Admittedly, these studies have not been designed to unearth the role of racism. The emerging, somewhat reluctant interpretation, is that racism is important. 5 39 Whittle concluded, “We believe that inadequate health education, differences in patients' preferences for invasive management, delivery systems that are unfriendly to members of certain cultures, and overt racism may all play a part.”44 Perhaps, as a legacy of racism, black patients distrust invasive diagnostic and therapeutic procedures and this inhibits them from seeking or accepting this type of care. In this climate of distrust, doctors may be inhibited in advising invasive procedures. If so, even if patients' preferences are partly responsible for the disparities, racism will not be wholly exonerated. The box shows my interpretation of current thinking on racism as a cause of health and healthcare disparities.

Interpretation of current thinking

  • Racial and ethnic inequalities in health and health care are abundant, but their underlying causes, and the contribution of racism, are controversial

  • Racial discrimination is evident in many factors that affect health, including employment and social security, and wealth inequalities

  • Minority groups find it difficult to reduce inequalities in wealth, partly because of racially discriminating actions and policies and because wealth may be accumulated over generations

  • Minority racial/ethnic groups generally (but not always) have worse overall health than the majority population

  • Health services are mostly staffed by members of the racial/ethnic majority, and are usually planned and delivered in relation to the needs and preferences of racial/ethnic majority users

  • Health services may offer a worse service to minority groups because staff treat patients unequally on account of their race or ethnicity, policies are based on the needs of the racial/ethnic majority, and specialist resources required to meet the needs of minority groups do not exist

  • That racial discrimination in health care is an obstacle to racial/ethnic minorities achieving their full health potential is sufficiently widely believed for it to be crucially important even without definitive evidence from research

Lessons from history and the US

In planning action amid controversy and uncertainty, we can draw upon two lessons from history and two from the United States. Firstly, our attitudes, interpretations, and actions are influenced by the prevailing ethos in society.15 As Gladys Reynolds candidly wrote “We the scientific community … bring everything we have been taught by our culture—our xenophobia, our homophobia, our racism, our sexism, our ‘classism,’ our tendency to ‘otherise’.”48 Most health workers and researchers are humanitarians, reflecting their professional ethos. This is an important but insufficient guard against racism.

Many scientists and policy makers of the 19th century shared the attitude that whites had a responsibility to colonise and lead coloured people, and perceived their actions as morally justifiable in the interest of society and all racial groups.1320 (Similar paternalistic views are illustrated in The Bell Curve, which argues that a consequence of a lower IQ in African Americans than white Americans is a set of specific policies attuned to that observation.50) The humanitarianism of medicine did not prevent doctors participating in Nazi medicine28 or in the unethical treatment of black South Africans in custody during the apartheid era. Doctors and other health professionals followed the ethos of their times.

Secondly, while important in showing inequities, seeking differentials in relation to race and ethnic group is potentially dangerous. Race science of the kind that dominated the 19th century is lurking, and The Bell Curve is an example, 50 51 but there are others. 52 53 Binet's test was designed to select children for “special educational attention,” but was used for immigration control and to show racial inferiority in intelligence.15 As Disraeli, the British prime minister, said to the House of Commons in 1849 “Race implies difference, difference implies superiority and superiority leads to predominance.”18 Research focusing on problems more common in minority groups and data presentation designed to highlight differences when minorities are compared with the majority population portray the minorities as weaker. 29 49 When published reports imply that genetic factors rather than environmental ones are the cause of racial differences in health, racial minorities may be perceived as innately weaker. 1 3 5 20 26 27 49

Two lessons from the United States are particularly noteworthy. Firstly, close and repeated observation and tracking of inequalities in this century has been accompanied by widening, not narrowing, of the gap between black and white people. Secondly, definitive answers on the role of racism as an explanatory factor for inequalities in health and health care have not emerged despite much debate, scholarship, and research. As a result, damaging allegations of racism in the health sector cannot be countered, and yet there is reluctance to take corrective action.

Conclusion: action amid uncertainty

We do not yet know to what degree racial and ethnic inequalities in health and health care are caused and maintained by racism. Should we wait to define the exact contribution of racism before we act or proceed according to the emergent (though uncertain) analysis that racism is important? Guinan has argued against delay, on the basis that we know enough.54 Warren, however, has strongly advocated that racism should be researched.55 The need and demand for more research was a dominant theme at the consultation on “Refining the research agenda” hosted by the National Centre for Health Statistics in May 1997 (personal observations).

There is understandable ambition to quantify the specific effects of racism, driven by such questions as “What are the causes of racial or ethnic disparities in health and health care?” and “What are the mechanisms by which the various causes operate?”56 Awaiting reliable quantitative answers to these complex questions will impede policy and action. Alternatively, we could simply accept that racial and ethnic inequalities result from a complex, inextricably linked set of factors, of which racism is an important part. The key question would then be “What actions could effectively reverse health and health care inequality?” The questions of cause and mechanism, including the role of racism, then become supportive of the quest for effective interventions.

The spectre of racism in healthcare institutions needs to be driven out. In the United Kingdom to date, the spotlight has mainly been on racism directed at ethnic minority staff.5759 With continuing disquiet about the potential adverse effects of racism on the quality of health care delivered to ethnic minority groups 10 21 22 25 60 61 and evidence (by self admission by the white population) of continuing racist attitudes in the United Kingdom,62 the spotlight is likely to shift to racism in patient care. Knowledge of events in the United States, and particularly the difficulties there in pinpointing the role of racism and in narrowing inequalities, may help to prevent the same happening in the United Kingdom. Knowledge of the history of race in science and medicine engenders the moral drive for action, provides insight on how societies acquire and interpret data on racial differences, shows how the prevailing social attitudes affect individuals' behaviour, and gives warning of the dangers to avoid.


I thank Drs Carl Shy, Trude Bennett, and Victor Schoenbach for critical and helpful comments on an earlier draft of this paper. I take sole responsibility for the ideas expressed. Much of the work in this paper was done while I was visiting professor in the Department of Epidemiology in the School of Public Health in the University of North Carolina at Chapel Hill, USA. I thank that institution, and particularly Drs D Savitz and C Shy, for hosting my visit and thank the University of Newcastle upon Tyne for granting me study leave.

Funding: No additional funding.

Conflict of interest: None.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
  54. 54.
  55. 55.
  56. 56.
  57. 57.
  58. 58.
  59. 59.
  60. 60.
  61. 61.
  62. 62.
View Abstract